Omnivision Technologies, Inc.

Black Sesame Technologies Inc.

Black Sesame International Holding Limited

Donghui Wu

Jizhang Shan

Chao Wang

Sarvesh Swami

Bin Chen

Tiejun Dai

Eiichi Funatsu

Jiangtao Kuang

Jiaju Yue

Zheng Yang

Xiao Xie

Timofey Uvarov

Chengming Liu

Xiaoyong Wang

Chen Hui Xu

Yaowu Mo

Jin Li

Guansong Liu

Min Qu

Jing Wang

Weihua Xiong

Xiaodong Luo

Ping-Hsu Chen

Zhaowei Xin

Lei Yang

Wenchao Zhang

Willie Teck Tian Png

Fangwei Cai

A method of image compression, including, receiving at least one unprocessed image frame, transforming a domain of the at least one unprocessed image frame to output a transformed domain dataset, block processing the transformed domain dataset to yield a blocked dataset, quantizing the blocked dataset to produce a quantized dataset and entropy encoding the quantized dataset to construct at least one compressed image frame.

Unprocessed image coding and decoding

An example method of single-exposure multiple-frame image capture, making an HDR video, comprising, capturing a series of frames at an underexposed setting, selecting a reference image from the series of frames, aligning the series of frames to the reference image, merging the series of frames into one frame based on the reference image, indexing the reference image to a next frame in the series of frames, resulting in an indexed reference image, realigning the series of frames to the indexed reference image and remerging the series of frames into a single frame based on the indexed reference image.

Generating high dynamic range video using low dynamic range image sensor

An example operation of depth map generation includes one or more of, simultaneously capturing a main-off camera image and an auxiliary-off camera image with an unpowered flash, sparse depth mapping an object based on the main-off camera image and the auxiliary-off camera image, capturing a main-on camera image with a powered flash, foreground probability mapping the object based on the main-off camera image and the main-on camera image and dense depth mapping the object based on the sparse depth map and the foreground probability map.

Multiple camera system with flash for depth map generation

A distributed light source couple-able to a mobile device, comprising, a distributed array of light emitting points and a mount connected to the distributed array and couple-able to the mobile device, the distributed array capable of light emission from a plurality of areas.

Flash array for portable camera system

A method of auto white balancing, including, receiving an original image, determining an RG logarithmic ratio of a set of red to green channel values of the original image, determining a BG logarithmic ratio of a set of blue to green channel values of the original image, determine an original two-dimensional histogram utilizing the RG logarithmic ratio and the BG logarithmic ratio, determine a Gaussian-blur two-dimensional histogram utilizing the RG logarithmic ratio and the BG logarithmic ratio, determining a sharpened two-dimensional histogram of a sharpened image utilizing the RG logarithmic ratio and the BG logarithmic ratio, determining a Laplacian-edge two-dimensional histogram of a Laplacian-edge image utilizing the RG logarithmic ratio and the BG logarithmic ratio and determining a white balancing gain utilizing a neural network based on the original 2D histogram, the Gaussian-blur 2D histogram, the sharpened 2D histogram and the Laplacian-edge 2D histogram.

Neural network based auto-white-balancing

A method of defect pixel correction, includes, receiving at least one center pixel signal in a plurality of frames, receiving a plurality of neighboring pixel signals adjacent the center pixel in the plurality of frames, determining a brightness of the center pixel signal, determining the brightness of the plurality of neighboring pixel signals, determining if the brightness of the center pixel signal exceeds a wounded pixel threshold of the plurality of neighboring pixel signals, determining a location of the center pixel having the brightness greater than the wounded pixel threshold in at least one frame, determining a number of reoccurrences of the center pixel having the brightness greater than the wounded pixel threshold, determining if the number of reoccurrences exceeds a defect pixel threshold and updating the at least one center pixel signal to a mean of the plurality of neighboring pixel signals.

Multiple frame defect pixel detection and correction

A method to reduce a motion blur and a rolling shutter effect, comprising, receiving a main image frame, a main timing and a main exposure from a main camera, receiving a secondary image frame, a secondary timing and a secondary exposure from a secondary camera, correcting the secondary image frame to the main image frame, determining a delta timing based on the main timing and the secondary timing, determining a delta exposure based on the main exposure and the secondary exposure, determining a discrete motion of offset sequences based on the secondary image frame, determining a row-wise motion blur kernel based on the main image frame and the discrete motion of offset sequences and determining a spatially varying kernel deconvolution based on the main image frame and the row-wise motion blur kernel.

Reducing global motion and rolling shutter in a dual camera system

A method of multiple-exposure multiple-frame image capture, comprising, capturing a first group of frames at an underexposed setting, selecting a first key frame from the first group of frames, aligning the first group of frames to the first key frame, merging the first group of frames into one first frame, capturing a second group of frames at the underexposed setting, selecting a second key frame from the second group of frames, aligning the second group of frames to the second key frame, merging the second group of frames into one second frame, selecting a main group of frames based on the first key frame and the second key frame, aligning the first key frame and the second key frame based on the main group of frames and merging the first key frame and the second key frame based on the main group of frames.

Two-stage method to merge burst image frames

A method of measuring lighting condition using dual cameras, having capturing a reference image, estimating a first minimum exposure target value, capturing a first plurality of minimum exposure value images, determining a first overall minimum exposure value, estimating a first maximum exposure target value, capturing a first plurality of maximum exposure value images, determining a first overall maximum exposure value of the first plurality of associated maximum exposure values and estimating a first dynamic exposure range as encompassing a first range having the first overall minimum exposure value as a first lower bound and the first overall maximum exposure value as a first upper bound.

Method of measuring light using dual cameras

A method for generating an HDR image includes (i) generating, from a plurality of single-exposure images of a scene, having a respective one of a plurality of exposure values, a sharpness map of sharpness values of the single-exposure images, and (ii) generating, from the plurality of single-exposure images, an exposure-value map of composite exposure values of the single-exposure images. The method also includes (iii) determining an optimal exposure value from the sharpness map and the exposure-value map, and (iv) generating the HDR image by combining a reference image of the scene, captured with the optimal exposure value, and at least two of the plurality of single-exposure images. An exposure selector for generating an HDR image, from the plurality of single-exposure images, includes a memory and a microprocessor. The memory stores non-transitory computer-readable instructions and is adapted to store the plurality of single-exposure images. The microprocessor is adapted to execute the aforementioned method.

Exposure selector for high-dynamic range imaging and associated method

A method of real-time depth mapping in a dual camera system, including, receiving an auto focus image, rectifying the auto focus image, receiving a plenoptic image having lens that cover 2&#xd7;2 pixels, extracting four plenoptic sub-images, averaging and rectifying the plenoptic sub-images, block matching the rectified auto focus image and the rectified plenoptic sub-images along epipolar lines, block matching the rectified plenoptic sub-images along a vertical and a horizontal direction, determining a composite block matching error of the plenoptic block match, determining a plenoptic disparity map of the plenoptic sub-images based on the composite block matching error, converting the plenoptic disparity map to a plenoptic depth map, converting the plenoptic depth map to an initial stereo disparity map, searching for a guided disparity map based on the initial stereo disparity map and the epipolar block match and converting the guided disparity map to a guided depth map.

Dual camera system for real-time depth map generation

A method for combining array camera images with feature-based ghost removal includes (a) receiving, from an array camera, a first image and a second image respectively captured by a first camera and a second camera of the array camera, (b) rectifying and aligning the first image and the second image, (c) after said rectifying and aligning, identifying features in the first image to produce at least one first feature image each indicating features in the first image, and identifying features in the second image to produce at least one second feature image each indicating features in the second image, (d) comparing the at least one first feature image with the at least one second feature image to determine a ghost mask defining combination weights for combination of the first image with the second image.

Array camera image combination with feature-based ghost removal

A lane detection system includes a non-volatile memory storing machine-readable instructions and an image processor capable of receiving a road image. The image processor, when executing the machine-readable instructions, is capable of: (i) processing the road image to identify a lane candidate within a lane-existing region of the road image, the lane-existing region having (a) a near subregion including an imaged road region nearer to the vehicle and (b) a far subregion including an imaged road region farther from the vehicle, (ii) verifying the lane candidate as a true lane candidate when a minimum distance between (a) a line fit to a portion of the lane candidate in the near subregion and (b) a predetermined reference point in the road image is less than a neighborhood distance; and (iii) extending the true lane candidate into the far subregion to form a detected lane marker demarcating the lane marker.

Lane detection system and method

A method of focusing an image sensor includes scanning a first portion of an image frame from an image sensor a first time at a first rate to produce first focus data. A second portion of the image frame from the image sensor is scanned at a second rate to read image data from the second portion. The first rate is greater than the second rate. The first portion of the image frame is scanned a second time at the first rate to produce second focus data. The first focus data and the second focus data are compared, and the focus of a lens is adjusted in response to the comparison of the first focus data and the second focus data.

Image sensor with fast intra-frame focus

A method for removing a ghost artifact from a multiple-exposure image of a scene method includes steps of generating and segmenting a difference mask, determining a lower threshold and an upper threshold, generating a refined mask, and generating a corrected image. The difference mask includes a plurality of absolute differences in luminance-values between the multiple-exposure image and a first image of the scene. The segmenting step involves segmenting the difference mask into a plurality of blocks. The lower and upper thresholds are based on statistical properties of the blocks. The method generates the refined mask by mapping each absolute difference to a respective one of a plurality refined values, of the refined mask, equal to a function of the absolute difference, the lower threshold, and the upper threshold. The corrected image is a weighted sum of the first image and the multiple-exposure image, weights being based on the refined mask.

Growing community of inventors

Sunnyvale, CA, United States of America

Donghui Wu